Re: [PATCH] Revert "iio: add modifiers for A and B ultraviolet light"

[Paul Cercueil <paul@crapouillou.net>]

Hi Javier,

Le mardi 23 janvier 2024 à 11:51 +0100, Javier Carrasco a écrit :
> On 23.01.24 11:43, Paul Cercueil wrote:
> > This reverts
> > b89710bd215e ("iio: add modifiers for A and B ultraviolet light")
> > 
> > Enum iio_modifer is *ABI*, you can't just decide to change all the
> > values from one version to another, otherwise you break userspace.
> > The new entries should have been added to the end of the enum.
> > 
> > Signed-off-by: Paul Cercueil <paul@crapouillou.net>
> > ---
> >  Documentation/ABI/testing/sysfs-bus-iio | 7 ++-----
> >  drivers/iio/industrialio-core.c         | 2 --
> >  include/uapi/linux/iio/types.h          | 2 --
> >  tools/iio/iio_event_monitor.c           | 2 --
> >  4 files changed, 2 insertions(+), 11 deletions(-)
> > 
> > diff --git a/Documentation/ABI/testing/sysfs-bus-iio
> > b/Documentation/ABI/testing/sysfs-bus-iio
> > index 2e6d5ebfd3c7..7937bb4a4a68 100644
> > --- a/Documentation/ABI/testing/sysfs-bus-iio
> > +++ b/Documentation/ABI/testing/sysfs-bus-iio
> > @@ -1587,8 +1587,6 @@
> > What:		/sys/.../iio:deviceX/in_intensityY_raw
> >  What:		/sys/.../iio:deviceX/in_intensityY_ir_raw
> >  What:		/sys/.../iio:deviceX/in_intensityY_both_raw
> >  What:		/sys/.../iio:deviceX/in_intensityY_uv_raw
> > -What:		/sys/.../iio:deviceX/in_intensityY_uva_raw
> > -What:		/sys/.../iio:deviceX/in_intensityY_uvb_raw
> >  What:		/sys/.../iio:deviceX/in_intensityY_duv_raw
> >  KernelVersion:	3.4
> >  Contact:	linux-iio@vger.kernel.org
> > @@ -1597,9 +1595,8 @@ Description:
> >  		that measurements contain visible and infrared
> > light
> >  		components or just infrared light, respectively.
> > Modifier
> >  		uv indicates that measurements contain ultraviolet
> > light
> > -		components. Modifiers uva, uvb and duv indicate
> > that
> > -		measurements contain A, B or deep (C) ultraviolet
> > light
> > -		components respectively.
> > +		components. Modifier duv indicates that
> > measurements
> > +		contain deep ultraviolet light components.
> >  
> >  What:		/sys/.../iio:deviceX/in_uvindex_input
> >  KernelVersion:	4.6
> > diff --git a/drivers/iio/industrialio-core.c
> > b/drivers/iio/industrialio-core.c
> > index 9a85752124dd..bce09d325142 100644
> > --- a/drivers/iio/industrialio-core.c
> > +++ b/drivers/iio/industrialio-core.c
> > @@ -117,8 +117,6 @@ static const char * const iio_modifier_names[]
> > = {
> >  	[IIO_MOD_LIGHT_GREEN] = "green",
> >  	[IIO_MOD_LIGHT_BLUE] = "blue",
> >  	[IIO_MOD_LIGHT_UV] = "uv",
> > -	[IIO_MOD_LIGHT_UVA] = "uva",
> > -	[IIO_MOD_LIGHT_UVB] = "uvb",
> >  	[IIO_MOD_LIGHT_DUV] = "duv",
> >  	[IIO_MOD_QUATERNION] = "quaternion",
> >  	[IIO_MOD_TEMP_AMBIENT] = "ambient",
> > diff --git a/include/uapi/linux/iio/types.h
> > b/include/uapi/linux/iio/types.h
> > index 5060963707b1..9c2ffdcd6623 100644
> > --- a/include/uapi/linux/iio/types.h
> > +++ b/include/uapi/linux/iio/types.h
> > @@ -91,8 +91,6 @@ enum iio_modifier {
> >  	IIO_MOD_CO2,
> >  	IIO_MOD_VOC,
> >  	IIO_MOD_LIGHT_UV,
> > -	IIO_MOD_LIGHT_UVA,
> > -	IIO_MOD_LIGHT_UVB,
> >  	IIO_MOD_LIGHT_DUV,
> >  	IIO_MOD_PM1,
> >  	IIO_MOD_PM2P5,
> > diff --git a/tools/iio/iio_event_monitor.c
> > b/tools/iio/iio_event_monitor.c
> > index 8073c9e4fe46..2eaaa7123b04 100644
> > --- a/tools/iio/iio_event_monitor.c
> > +++ b/tools/iio/iio_event_monitor.c
> > @@ -105,8 +105,6 @@ static const char * const iio_modifier_names[]
> > = {
> >  	[IIO_MOD_LIGHT_GREEN] = "green",
> >  	[IIO_MOD_LIGHT_BLUE] = "blue",
> >  	[IIO_MOD_LIGHT_UV] = "uv",
> > -	[IIO_MOD_LIGHT_UVA] = "uva",
> > -	[IIO_MOD_LIGHT_UVB] = "uvb",
> >  	[IIO_MOD_LIGHT_DUV] = "duv",
> >  	[IIO_MOD_QUATERNION] = "quaternion",
> >  	[IIO_MOD_TEMP_AMBIENT] = "ambient",
> Oops, sorry about that. You are right, I will send a new patch with
> the
> new elements at the end of the enum. This patch should be applied to
> get
> things right again, though.

Np.

I notice now that we can't really apply the revert as the veml6075
driver is already using these enum values - so applying it would result
in build errors.

Can you just move these entries to the end of the enum in your new
patch?

> 
> Thanks and best regards,
> Javier Carrasco

Cheers,
-Paul